2

Assume the following summation,

$$ \sum_{i=0}^{1000}\left(-1\right)^{i}{1000 \choose i}\left(100 - i\right)^{500}. $$

I know that this summation is zero, $0$ ( I've checked it with Maple, though ). But I cannot find any proof for that!. Can you provide any help ?.

P.S. This is not a homework problem, just a question one of my friend asked me.

Felix Marin
  • 94,792
Ali
  • 583
  • Split into summing even and odd series and reduce maybe? – Chinny84 Mar 03 '14 at 11:24
  • @Chinny84 I've done this, even I've expanded it using binomial theorem which gives me two nested sums. None of them helped me (I am not so good in counting and combinatorics) – Ali Mar 03 '14 at 11:32

2 Answers2

1

If $(Δ p)(x)=p(x+1)-p(x)$ denotes the step-one difference operator, your expression is equal to $$0=(Δ^{1000} p)(x)\quad\text{where}\quad p(x)=(100-x)^{500}$$

Since each application of $Δ$ lowers the degree of the polynomial by one, already $(Δ^{501} p)(x)=0$, and moreso the higher order differences.


If $p(x)=x^k$, then $$ (Δ p)(x)=p(x+1)-p(x)=(x+1)^k-x^k=(x+1)^{k-1}+(x+1)^{k-2}x+...+(x+1)x^{k-2}+x^{k-1} $$ is a polynomial of one degree less, this extends linearly to all polynomials. Thus $\deg(Δ^m p)=\deg p-m$ as long as $m\le\deg p$, and $Δ^{1+\deg p} p=0$.

For the higher order differences on gets the formula $$ (Δ^m p)(x)=\sum_{i=0}^m(-1)^{m-i}\binom{m}{i} p(x+i) $$ This is obviously true for $m=0$, and easy to check for $m=1$, let's do the induction step \begin{align} (Δ^{m+1} p)(x)&=(Δ^m p)(x+1)-(Δ^m p)(x) \\ &=\sum_{i=0}^m(-1)^{m-i}\binom{m}{i} p(x+i+1)-\sum_{i=0}^m(-1)^{m-i}\binom{m}{i} p(x+i) \\ &=\sum_{i=1}^{m+1}(-1)^{m-i+1}\binom{m}{i-1} p(x+i)+\sum_{i=0}^m(-1)^{m-i+1}\binom{m}{i} p(x+i) \\ &=\sum_{i=0}^{m+1}(-1)^{m+1-i}\binom{m+1}{i} p(x+i) \end{align}


Even easier is the derivation using the translation operator $(Tp)(x)=p(x+1)$. Then $Δ=(T-I)$ and $$ Δ^m=(T-I)^m=\sum_{i=0}^m\binom{m}{i}(-1)^{m-i}T^{i}. $$

Lutz Lehmann
  • 132,997
  • Sorry, I cannot get the intuition of your argument. Would you mind explaining this further? – Ali Mar 03 '14 at 11:35
0

$\newcommand{\+}{^{\dagger}} \newcommand{\angles}[1]{\left\langle #1 \right\rangle} \newcommand{\braces}[1]{\left\lbrace #1 \right\rbrace} \newcommand{\bracks}[1]{\left\lbrack #1 \right\rbrack} \newcommand{\ceil}[1]{\,\left\lceil #1 \right\rceil\,} \newcommand{\dd}{{\rm d}} \newcommand{\down}{\downarrow} \newcommand{\ds}[1]{\displaystyle{#1}} \newcommand{\equalby}[1]{{#1 \atop {= \atop \vphantom{\huge A}}}} \newcommand{\expo}[1]{\,{\rm e}^{#1}\,} \newcommand{\fermi}{\,{\rm f}} \newcommand{\floor}[1]{\,\left\lfloor #1 \right\rfloor\,} \newcommand{\half}{{1 \over 2}} \newcommand{\ic}{{\rm i}} \newcommand{\iff}{\Longleftrightarrow} \newcommand{\imp}{\Longrightarrow} \newcommand{\isdiv}{\,\left.\right\vert\,} \newcommand{\ket}[1]{\left\vert #1\right\rangle} \newcommand{\ol}[1]{\overline{#1}} \newcommand{\pars}[1]{\left( #1 \right)} \newcommand{\partiald}[3][]{\frac{\partial^{#1} #2}{\partial #3^{#1}}} \newcommand{\pp}{{\cal P}} \newcommand{\root}[2][]{\,\sqrt[#1]{\vphantom{\large A}\,#2\,}\,} \newcommand{\sech}{\,{\rm sech}} \newcommand{\sgn}{\,{\rm sgn}} \newcommand{\totald}[3][]{\frac{{\rm d}^{#1} #2}{{\rm d} #3^{#1}}} \newcommand{\ul}[1]{\underline{#1}} \newcommand{\verts}[1]{\left\vert\, #1 \,\right\vert}$ $\ds{\sum_{i=0}^{1000}\pars{-1}^{i}{1000 \choose i}\pars{100 - i}^{500} = 0:\ {\large ?}}$

\begin{align} &\sum_{i=0}^{1000}\pars{-1}^{i}{1000 \choose i}\pars{100 - i}^{500} =\sum_{i=0}^{1000}\pars{-1}^{i}{1000 \choose i}\sum_{\ell = 0}^{500} {500 \choose \ell}100^{500 - \ell}\pars{-i}^{\ell} \\[3mm]&=\sum_{\ell = 0}^{500}{500 \choose \ell}\pars{-1}^{\ell}100^{500 - \ell} \underbrace{\sum_{i=0}^{1000}{1000 \choose i}\pars{-1}^{i}i^{\ell}} _{\ds{=\ 0;\quad\mbox{See bellow}}} \end{align}

\begin{align} \pars{1 - z}^{1000}&=\sum_{i = 0}^{1000}{1000 \choose i}\pars{-1}^{i}z^{i} \\ -1000z\pars{1 - z}^{999}&=\sum_{i = 0}^{1000}{1000 \choose i}\pars{-1}^{i}iz^{i} \\ -1000z\pars{1 - z}^{999} + 1000\times999 z^{2}\pars{1 - z}^{998}& =\sum_{i = 0}^{1000}{1000 \choose i}\pars{-1}^{i}i^{2}z^{i}\,, \end{align} If we continue this procedure we get increasing powers of $i$ ( up to $i^{\ell}$ with $\ell \leq 500$ ) in the right member. Then, set $z = 1$ in both members and we'see that all the sums vanish out.

Felix Marin
  • 94,792